Welcome to the documentation of Spectra analyzer. This is a WebSocket based web application built on Flask and SocketIO API frameworks. You can use it to simply download spectra from virtual observatory archives using protocols SSAP and DataLink. You can also use this tool to inspect the spectra using dimension reduction method continual wavelet transformation.
This tool is tightly coupled with the Spectra downloader tool that is capable of downloading spectra from virtual observatory archives using protocols SSAP and DataLink. Both repositories were created as a semestral project for subject MI-PYT at FIT CTU in Prague. For more information about Spectra downloader refer to its repository.
Documentation can be found online on spectra-analyzer.readthedocs.io or you can build it manually by clonning the repository and executing:
cd docs python3 -m pip install -r requirements.txt make html
Documentation is now created in
Semestral work task
The aim of this work is to create a tool for analyzing a dimensionality reduction of a chosen astronomical spectrum.
The tool has a web interface written using Python framework Flask. Astronomical spectra can be downloaded from the Virtual Observatory archive using specialized astronomical protocols SSAP and Datalink. The principle of these protocols is to query a specific URL address using HTTP protocol in order to obtain spectrum itself or XML formatted file containing information about a construction of additional URL addresses. The tool uses the Python library requests for fetching a content from the specified URL address and the library xml.sax to parse XML documents. For downloading either multiple threads or the Python asyncio library should be used.
Any spectrum either downloaded by the tool or already present in the filesystem can be chosen for further inspection. The tool visualize the selected spectrum and it also allows user to change parameters of dimensionality reduction method continual wavelet transformation. The transformed spectrum is also visualized and a user can observe changes between the original and the transformed spectrum. The task of this tool is to allow a user to find out a transformation parameters threshold where the spectrum still has the most of an original important information but its size is reducted by the dimensionality reduction method. The tool uses Python modules matplotlib, numpy and mply for the spectrum manipulation and transformation.